The structure of a graph is made up of nodes (also known as vertices) and edges. Nodes denote points in the graph data. For example, accounts, customers, devices, groups of people, organizations, products or locations may all be represented as a node. Edges symbolize the relationships, or lines of communication, between nodes. Every edge can have a direction, either one-way or bidirectional, and a weight, to depict the strength of the relationship.
Once the graph database is constructed, analytics can be applied. The algorithms can be used to identify values or uncover insights within the data such as the average path length between nodes, nodes that might be outliers and nodes with dominant activity. It can also be used to arrange the data in new ways such as partitioning information into sections for individual analysis or searching for nodes that meet specific criteria.
Types of graph analytics
There are four main types of analytics that can be applied to graphs:
- Path analysis- This focuses on the relationships between two nodes in a graph. This type of graph analytics can help identify the shortest path between nodes, find the widest path between weighted nodes and calculate a spanning tree around a center point.
- Connectivity analysis- This focuses on the weight of the edges between nodes. It can be applied to identify weaknesses in a system or anomalies such as abnormally high or low activity.
- Community analysis- This focuses on the interactions between nodes. It clusters nodes into labeled groups of similar objects to help with organization.
- Centrality analysis- This focuses on the relevancy of each node in a graph. It can be used to rank popularity or influence between nodes.
Examples of applications for graph analytics
Graph analytics can be used for a variety of applications, such as:
- Detecting cybercrimes such as money laundering, identity fraud and cyberterrorism.
- Applying analysis to social networks and communities such as monitoring statistics and identifying influencers.
- Performing analysis on the traffic and quality of service for computer networks.
- Optimizing logistics for manufacturing and transportation industries.
- Determining page rank analytics and tracking their popularity or amount of clicks.
- Analyzing the parts of a software application and how they interact to find potential issues.